creating-agents-in-medusa

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill intentionally exposes a MedusaExec facility that writes and executes arbitrary TypeScript with full DI container/database access and streams results to an external model, creating clear remote-code-execution and data-exfiltration risks if the admin-only boundaries are ever misconfigured or bypassed.